Space-Time-Frequency Analysis of EEG Data in Semantic Inhibition: Control of False Positives in Single Subjects

After testing for differences between task conditions in single subjects, principal component analysis (PCA) was applied sequentially to spectral, spatial, and temporal dimensions, in that order, keeping subjects as samples. Results The statistical distribution of the log power spectra in null regions of the spectrograms were normally distributed, so p values at each pixel were computed using parametric tests (Bokil et al., 2007).